Skip to main content

Table 2 Relationship of Global Fund support to adult and under-five mortality changes

From: Mortality changes after grants from the Global Fund to Fight AIDS, tuberculosis and malaria: an econometric analysis from 1995 to 2010

 

All-cause adult mortality

All-cause under-five mortality

Malaria-specific child mortality

(p-value)5

(p-value)5

(p-value)5,6

Global Fund $ per capita1

−0.0014 (0.005)

−0.0005 (0.33)

−0.0069 (0.033)

Global Fund $ per capita × HWD2

0.0013 (0.72)

−0.0005 (0.25)

0.0016 (0.40)

Health workforce density (HWD) 2

−0.0021 (0.090)

−0.0022 (0.11)

−0.0094 (0.15)

GDPpc (logged, in 2005 USD, PPP adjusted)3

−0.0587 (0.16)

−0.1484 (0.067)

−0.6625 (0.032)

% of urban population

0.0010 (0.77)

−0.0050 (0.23)

0.0178 (0.54)

Health expenditure per capita (logged, in 2005 USD)4

−0.0018 (0.92)

−0.0078 (0.60)

−0.0503 (0.55)

Number of countries

147

147

55

Number of country-year observations

2322

2322

856

  1. Notes:
  2. 1Global Fund disbursements per capita in constant 2005 USD. A significant negative coefficient indicates effectiveness of Global Fund in bending down the mortality trend. The coefficient indicates that the mortality rate declined by coefficient × 100 percent faster per year for every $1 increase in per capita disbursements
  3. 2HWD is health workforce density, defined as the number of doctors and nurses per 100,000 population, an indicator of health system capacity. All HWD estimates were log-transformed. The interactions between HWD and the Global Fund variables indicate whether, at any level of Global Fund exposure, the observed outcomes changed based on HWD. The coefficient of the interactive term measures how the health workforce density in a country modifies the effectiveness of Global Fund in changing the mortality trend. There is no evidence that the effect of Global Fund on mortality was meaningfully different based on health system capacity as measured through HWD
  4. 3Gross domestic product per capita, log-transformed, in 2005 USD, adjusted for purchasing power parity
  5. 4Total health expenditures minus Global Fund disbursements from all sources per capita in 2005 USD
  6. 5Mortality rates are log-transformed, so that the coefficient values can be interpreted as the additional annual proportional change in mortality with each additional year of support from the Global Fund (see numerical examples in Results text)
  7. 6The malaria-specific models analyze countries with high, severe or extreme malaria burden. According to the Global Fund’s eligibility criteria, these are the countries with burden defined as high, severe, or extreme; countries with low or moderate burden were excluded, resulting in the analysis of 55 countries. Only malaria grants were used to examine malaria-specific mortality
Back to article page

BMC Public Health

ISSN: 1471-2458

Contact us